Modern computing devices, such as mobile phones, tablets, and laptops, provide capabilities to capture images. Still image capture functionality during video recording with these devices may be performed by extracting a single frame out of the video sequence. However, the exposure time for a single video frame is limited by the video frame rate, resulting in a still image having a lower image quality.
Multiple image frames from a video sequence may be used to generate a higher quality still image by combining those multiple frames, such as multi-frame noise filtering or super-resolution processing. While a higher quality still image is generated with such systems, devices that are equipped with optical image stabilization (OIS) attempt to keep the image steady over the entire sensor exposure to get less motion into the frame and to stabilize the consecutive image frames to achieve smoother end results. Because there is movement between the consecutive frames, when OIS is performed, image alignment of the captured image frames is needed before the image frames can be combined. This image alignment process results in an increased processing load, increased use of memory and/or increased battery power consumption.